TITLE:
Analysis and Interpretation of EEG Signals for the Design of an Automated Epilepsy Detection through an Ensemble of Intelligent Systems
AUTHORS:
Romain Atangana, Amstrong Emini Me Zenanga, Daniel Tchiotsop, Godpromesse Kenne
KEYWORDS:
Epilepsy Seizure, EEG Signals, Intelligent System, Discrete Wavelet Transform (DWT), Schapire Boosting Technique, ROC Curves, Confusion Matrix
JOURNAL NAME:
World Journal of Neuroscience,
Vol.16 No.3,
June
30,
2026
ABSTRACT: Epileptic seizures are characterized by abnormal neuronal discharges that cause significant disturbances in brain electrical activity. Electroencephalography (EEG) is widely used for epilepsy diagnosis, but manual interpretation of EEG signals is time-consuming and prone to human error. In this study, we propose an automated epilepsy detection system based on an ensemble of intelligent classifiers. The proposed framework first employs Discrete Wavelet Transform (DWT) to decompose EEG signals into five frequency sub-bands (Delta, Theta, Alpha, Beta and Gamma). Linear Discriminant Analysis (LDA) is then applied for feature dimensionality reduction. Three machine learning classifiers, namely Multilayer Perceptron (MLP), Support Vector Machine (SVM) and K-Nearest Neighbors (KNN), are combined using a majority voting strategy within a multi-agent boosting framework. Experiments were conducted on the publicly available University of Bonn EEG dataset. The proposed ensemble system achieved a classification accuracy of 100% with a significantly reduced global interpretation error compared to individual classifiers. These results demonstrate the effectiveness of the proposed intelligent ensemble approach for assisting neurologists in epilepsy diagnosis.